# Ev is not the measure of everything, and not the unit of anything

#### Doug Kerr

##### Active member
For many years, we have used the unit "stop" in the "logarithmic" expression of every quantity that has a role in the matter of photographic exposure. Those quantities include (but are not limited to):

Direct quantities

• Aperture (or effective aperture)
• Exposure time ("shutter speed")
• Imaging sensitivity (e.g., ISO speed)
• Scene luminance

Indirect quantities

• Amount of exposure compensation

Of course a change in one of the quantities of "one stop" represents a 2:1 or 1:2 change in the actual underlying quantity (in the case of aperture that being the area, not the diameter).

Now "stop" is not a unit that has an "elegant" background. So in modern times, many authors have tried to replace it with something more "elegant". And sadly, that is often, and mistakenly, the "unit" "Ev".

Ev is not actually a unit; it is a quantity, and is expressed in dimensionless, unitless numbers - just "1", "2", and so forth.

It is a quantity defined by APEX, the Additive System of Photographic Exposure, and expresses, in base-2 logarithmic terms, the joint effect on photographic exposure of a certain aperture (in terms of an f-number) and a certain exposure time. Period. It has no other legitimate use.

So we can say, for example, if our trial shot was a little bit overexposed, we may wish to decrease the photographic exposure by a factor of 1:2 (a "one stop" decrease, in traditional terms); that is, increase the Ev by one unit. Increase? Yes, Ev is defined in that direction, a greater value representing less photographic exposure. But we don't say, "decrease the photographic exposure by one Ev".

Now suppose we just want to decrease our exposure time by a factor of 1:2. Can that properly be expressed as a change of one in the Ev?

Well, if we don't also change the aperture, that is true. But the exposure time itself (independent of its conspiracy with aperture) is properly stated as the quantity Tv, the APEX logarithmic quantity for exposure time. And of course the unit of Tv is not "the TV"; line Ev, Tv is expressed in dimensionless, unitless numbers.

Now we get to the sensitivity of a camera (perhaps the "ISO speed" or the "ISO SOS"). If we double that, have we made a change of "one Ev"? No, for two reasons:

• Ev only applies to photographic exposure, that is, the joint effect of an aperture and an exposure time. It does not apply to camera sensitivity, or the temperature of one's coffee, or the average wage of a photographer's assistant.

• It any case, "the Ev" is not a unit. We can speak of changing exposure, expressed as EV, by one unit, not changing it by "one Ev".

Now what about exposure compensation? On many cameras, the control for this is denominated in "Ev". We may read that if our trial metered image is a little too bright, we might want to apply "an exposure compensation of -1 Ev".

That is not correct, because:

• Ev only applies to photographic exposure, that is, the joint effect of an aperture and an exposure time.

• It any case, the Ev is not a unit. We can speak of changing exposure, expressed as EV, by one unit, not changing it by "one Ev".

Now, it could be argued that, because a one-unit change in EC should lead to a one-unit change in the metered Ev, that it is proper in the case of EC to think of that control as being a control of Ev (on a relative basis). But in any case, a one unit change in Ev (or EC) is not a "one Ev change". It is a change of one in the Ev (or EC)

So, when it comes for a universal "unit" of change in these various quantities, "stop", ragged as its history is, is the only one that is not, in so many cases, just plain "wrong".

Best regards,

Doug

#### Maris Rusis

##### Member
Interesting! I reckon the Ev concept is another example of photographic technology trying to make things simpler at the cost of actually creating ambiguity and confusion. Ok, if you square the lens aperture (expressed as an f number) and divide that by the shutter speed (measured in seconds) and then take the logarithm (to base 2) of the result you get Ev.

But, and this is where I get tricked up, the Ev number only describes the setting of a couple camera controls, aperture and shutter speed, and doesn't describe exposure as such. That depends not only on Ev but subject luminance as well. All I know is that any combination of apertures and shutter speeds that give me the same Ev will give me the same exposure if the subject luminance doesn't change. Whether this unchanging exposure is right or wrong ultimately depends on how I set the camera using a well calibrated light meter.

And a well calibrated light meter is no trivial thing if I read The Pumpkin correctly.

#### Doug Kerr

##### Active member
Hi, Jerome,

In that Wikipedia passage, it would have been better to say:

Strictly, EV is not a measure of luminance or illuminance. However, EV is often used to indicate the luminance (or illuminance) for which a camera with a given ISO speed would use the indicated EV to obtain the nominally correct exposure. In particular, it is common practice among photographic equipment manufacturers to express luminance in EV for ISO 100 speed, as when specifying metering range or autofocus sensitivity.​

Best regards,

Doug

#### Robert Watcher

##### Active member
It's a good thing people are around to educate about this stuff for those with whom it matters.

Funny thing is that I have been shooting professionally for 35 years and although I've seen the term EV in all kinds of photo mags and sales literature - I have never had an interest in knowing what a EV is. Like all good technical conversations about photography, lack of my understanding what is being talked about has had absolutely 0 effect on my ability to produce photographs that I am happy with and that have paid me very well.

But as I have told you before Doug, I think it is nice to have well rounded participation on the forum, and there certainly aren't many people left who will put the effort and dedication into these scientific findings and writings. I find that I learn something new every now and then when I take the time to peruse such a post as this. My simple, visual mind may still be confused on this matter however. lol thumbsup

#### Doug Kerr

##### Active member
Hi, Rob,

It's a good thing people are around to educate about this stuff for those with whom it matters.

Funny thing is that I have been shooting professionally for 35 years and although I've seen the term EV in all kinds of photo mags and sales literature - I have never had an interest in knowing what a EV is. Like all good technical conversations about photography, lack of my understanding what is being talked about has had absolutely 0 effect on my ability to produce photographs that I am happy with and that have paid me very well.

But as I have told you before Doug, I think it is nice to have well rounded participation on the forum, and there certainly aren't many people left who will put the efgort and dedication into these scientific findings and writings. I find that I learn something knew every now and then when I take the time to peruse such a post as this. My simple, visual mind may still be confused on this matter however. lol thumbsup
Thank you (I think).

Best regards,

Doug

#### Robert Watcher

##### Active member
Hi, Rob,

Thank you (I think).

Best regards,

Doug
It's not a slam. I'm being supportive believe it or not. If your posts weren't here, I wouldn't even think about such subjects. So you make me curious and I read what you have to say.

#### Doug Kerr

##### Active member
Hi, Rob,

It's not a slam. I'm being supportive believe it or not. If your posts weren't here, I wouldn't even think about such subjects. So you make me curious and I read what you have to say.
I know! I was just pulling your leg.

Thanks so much for your support.

Best regards,

Doug

#### Doug Kerr

##### Active member
Hi, Maris,

Interesting! I reckon the Ev concept is another example of photographic technology trying to make things simpler at the cost of actually creating ambiguity and confusion.
Only when it is abused!

Ok, if you square the lens aperture (expressed as an f number) and divide that by the shutter speed (measured in seconds) and then take the logarithm (to base 2) of the result you get Ev.
Well said.

But, and this is where I get tricked up, the Ev number only describes the setting of a couple camera controls, aperture and shutter speed, and doesn't describe exposure as such.
Part of what's likely bothering you is that there are two (at least) meanings of "exposure":

Photographic exposure is the joint effect of an aperture and a shutter speed. That is,of course, what Ev is the logarithmic expression of.

Photometric exposure is the product of the illuminance on the film/sensor and the exposure time. (These days, it is represented by the symbol H.)

That [that is, photometric exposure] depends not only on Ev but subject luminance as well. All I know is that any combination of apertures and shutter speeds that give me the same Ev will give me the same exposure if the subject luminance doesn't change.
Indeed.

Whether this unchanging exposure is right or wrong ultimately depends on how I set the camera using a well calibrated light meter.
And of course on your criteria for "right", there being no mathematically-determinable definition of what is the "right" exposure result (a third use of the word exposure).

And a well calibrated light meter is no trivial thing if I read The Pumpkin correctly.
Indeed!

Thanks.

Best regards,

Doug

#### Asher Kelman

##### OPF Owner/Editor-in-Chief
So, Doug, while I've been busy dealing with broken A/C and no hot water, you have pretty well scarpered off with the Ev rating which I thought was as reliable as an Enfield Rifle or as the republicans disavowing science.

Ev then seems then like the words "freedom", "sex" and "rights" that we use a lot and generally think we understand, but are actually mostly oblivious to some exact meaning.

Asher

#### Doug Kerr

##### Active member
Hi, Asher,

So, Doug, while I've been busy dealing with broken A/C and no hot water, you have pretty well scarpered off with the Ev rating which I thought was as reliable as an Enfield Rifle or as the republicans disavowing science.

Ev then seems then like the words "freedom", "sex" and "rights" that we use a lot and generally think we understand, but are actually mostly oblivious to some exact meaning.
Or "patriotism". Or "constitutional". One politician said the other day that a recent decision of the Supreme Court was "unconstitutional". I'm not sure what part of the Court's role (since Marbury vs. Madison) he didn't understand.

Good luck with the a/c and the water heater. My domestic engineering of late has been limited to the dishwasher!

Best regards,

Doug

#### Jerome Marot

##### Active member
In that Wikipedia passage, it would have been better to say:

Strictly, EV is not a measure of luminance or illuminance. However, EV is often used to indicate the luminance (or illuminance) for which a camera with a given ISO speed would use the indicated EV to obtain the nominally correct exposure. In particular, it is common practice among photographic equipment manufacturers to express luminance in EV for ISO 100 speed, as when specifying metering range or autofocus sensitivity.​
Probably. But what I wanted to point out is that, in practice, EV is actually related to lux (the unit). I have an old EV meter, of the type which uses a photo-cell and a small voltmeter and no battery. Its scales are labelled in EV. I have another one very similar device, but which is labelled in lux. I believe that there are such meters which have actually both scales on the same dial.

Certainly, if we have a physical device used to mesure both units with a simple change of the dial, the two units are somewhat related.

#### Doug Kerr

##### Active member
Hi, Jerome,

Probably. But what I wanted to point out is that, in practice, EV is actually related to lux (the unit). I have an old EV meter, of the type which uses a photo-cell and a small voltmeter and no battery. Its scales are labelled in EV.
If this did not have an exposure index (ISO) setting, it was probably intended for use with a camera with a fixed ISO speed (for example a Polaroid camera of a certain generation) and equipped with a shutter having Ev denomination for combinations of aperture and shutter speed. If the situation is actually as I conjectured, this of course is a perfectly legitimate use of Ev (to describe a photographic exposure).

I have another one very similar device, but which is labelled in lux.
Ah, presumably an incident light meter.

I believe that there are such meters which have actually both scales on the same dial.
Many exposure meters have an "output" reading in Ev (meaning that its recommendation for photographic exposure is delivered as a value of Ev in addition to a continuum of valid aperture/shutter speed values). This of course is a perfectly legitimate use of Ev (to describe a photographic exposure).

Certainly, if we have a physical device used to mesure both units with a simple change of the dial, the two units are somewhat related.
Indeed. Of course if the instrument observes either luminance or illuminance, then its "calculator dial" (having been set with an exposure index in terms of ISO speed) will essentially solve the exposure equation and can thus quite properly yield a "result" as Ev. So, with some lack of rigor, we might say that this meter "measures" Ev.

Thanks.

Best regards,

Doug

#### Doug Kerr

##### Active member
Hi, Jerome,

Here we see the latest (and likely last) of the "Norwood Director" line of exposure meters, the Sekonic L-389A (Studio Deluxe III):

Douglas A. Kerr: Sekonic L-398A exposure meter

This instrument is primarily intended for incident light measurement (although it can be equipped with a different "input organ" suiting it for reflected light measurement).

The meter proper delivers its reading on a scale denominated in footcandles (a measure of illuminance). That scale is the premise for the transfer of the reading to the calculator dial system. (If using the meter in the "reflected light" mode, this can just be considered to be an arbitrary number.)

The calculator dial system has a setting for exposure index, in terms of ISO speed, read in a small window.

The "result" is a recommended photographic exposure, delivered in two ways:

• As a continuum of aperture-shutter speed combinations, read along adjacent scales on two of the dials. In the picture, one such combination for the assumed measurement would be 1/30 sec and f/5.6.

• As an Ev value, read in the window seen at about "11 o'clock" on the calculator (in the picture, Ev 9.9, which corresponds to, for example, 1/30 sec and f/5.6).

Ev is here used in its proper sense, a single designator of photographic exposure, the effect of a certain aperture and a certain exposure time (shutter speed) (in this case, the one "recommended" by the exposure meter).

This instrument is from our extensive collection of "Norwood Director" exposure meters.

By the way, the secondary marking on the scale of the meter proper, a series of f-number values (red), is part of the Norwood "direct reading" system. To use this, an attenuator (a perforated metal plate) is inserted, in a slot, between the hemispherical collector and the photodetector itself. The attenuator, taken from a set provided with the meter (or, for some models, purchased separately), is marked with one or more combinations of ISO speed and shutter speed, and is chosen to suit those particulars of the upcoming shot.

This having been done, the recommended aperture (as an f-number) is read directly from the meter scale.

This mode was especially convenient in cinematography (the original intended market for this series of meters) where the shutter speed was ordinarily fixed (that being a creature of the frame rate - ordinarily fixed - and the shutter angle, normally fixed for any given camera) and the ISO speed (for a particular type of film stock) would ordinarily be constant for a substantial series of shots.

Best regards,

Doug

#### Doug Kerr

##### Active member
At one time it was popular on cameras to include provisions for setting photographic exposure in terms of Ev.

Here we see a Prontor SVS shutter, size 0, with provisions for indicating and setting exposure in terms of Ev. It is on my Graflex Century Graphic press camera (2¼”×3¼” format size).

Please excuse the bad image. It was done a number of years ago from an irregular source image.​

This feature was introduced by Prontor Werk in 1958.

The Ev coupler lever determines whether or not, as the shutter speed dial is turned, the aperture setting also changes to maintain the Ev shown (or set) by the Ev indicator. The coupler is disengaged if the photographer wishes to set shutter speed and aperture independently and just be able to see the resulting Ev. It is engaged if the photographer wishes to “set” the exposure in terms of Ev and play the tradeoff between shutter speed and aperture by simply moving the shutter speed dial.

Best regards,

Doug

#### Doug Kerr

##### Active member
Earlier in this thread, Jerome referred to the convention in which an "Ev" value is used to denote a scene luminance, namely the luminance which, under the "standard" exposure equation, would cause the camera to adopt a photographic exposure of the stated Ev. He went on to point out that of course for this to be meaningful, the exposure index (ISO speed) that was assumed needed to be stated, although in fact the usual convention was for that to be ISO 100.

And that is in fact quite true.

He then pointed out that to be rigorous, the assumed metering constant (a parameter of the actual exposure equation used by the camera) should also be stated.

And that is in fact quite true. But the way that plays out is curious indeed.

If we are working with APEX quantities (the base-2 logarithmic system of expressing various values of concern in the area of photographic exposure, Ev being one member of that system) we find that the exposure equation is:

Ev = Bv + Sv​

where Ev represents the photographic exposure (the joint effect of a certain aperture and a certain shutter speed), Bv represents the metered scene luminance, and Sv represents the exposure index as an ISO speed.

Notably missing from this is any recognition of the reflected light metering constant, K. Yet the context in which the APEX system was constructed recognized and honored the doctrine that the value of that metering constant could be chosen by the camera manufacturer (over a modest range) based on its judgement of what would, overall, give the best overall exposure results for its customers.

And in the basic (non-logarithmic) form of the standard exposure equation one term is in fact the metering constant.

But we don't see it in the logarithmic (APEX) form of the metering equation.

Well, the answer to this riddle is that the APEX value Bv, which represents scene luminance, does not have a fixed definition only in terms of luminance, L. Rather, its definition includes the value of K adopted by the camera (or meter) manufacturer.

Oh, great!

Why was this done? To make the APEX form of the exposure metering equation simple, since it was originally thought that this would be used by photographers to make exposure decisions.

But if the photographer started with, for example, an assumed value of Bv for some situation, taken from a table, under what value of K would that value have been reckoned? Well, it of course doesn't really matter, since these are very broad estimates anyway!

And suppose that in some scientific or engineering work we wish to speak of luminance in terms of Bv. This work does not involve any particular camera (or meter). Well, we just have to pick a handy value of K (and disclose it in our report).

But let's get back to the original story. If we know that the predicate of the use of "Ev" to denote scene luxuriance is ISO 100 (Sv 5), then we can get the luminance that implies, expressed as Bv (the legitimate APEX value for luminance) very easily, this way (I will use Ev to mean "Ev" as an expression of luminance):

Bv = Ev - Sv, thus
Bv = Ev -5

And no value of the metering constant is needed!

But the rub is that this value of Bv does not mean (precisely) a certain value of luminance. To get that, we need to take into account the applicable value of K.

But again, as in the other situation I described above, it really doesn't matter. If we have been told that our camera can do reliable AF for scene luminance down to "Ev 8" (which means Bv 3, considering the predicate of ISO 100), we don't treat that as precise within a few percent. (If we had such a precise answer, what would we do with it?) So the value of K upon which that value of Bv is defined is not of any real concern to us.

Best regards,

Doug